Materials Transactions, Vol. 54, No. 12 (2013) pp. 2209 to 2214
© 2013 The Japan Institute of Metals and Materials
Effect of Alloying Elements with Positive Heat of Mixing on the Free Volume
and Compressive Plasticity in ZrCoCuAl Bulk Metallic Glasses
Z. Liu1, K. C. Chan1,+ and L. Liu2
1
Department of Industrial and Systems Engineering, The Hong Kong Polytechnic University, Hong Kong, China
The State Key Lab of Die and Mould Technology, Huazhong University of Science and Technology, 430074 Wuhan, China
2
In this work, the effect of alloying an element with positive heat of mixing with Co on the free volume and compressive plasticity of a
ZrCoAl bulk metallic glass was investigated. By substituting Co with Cu, fully amorphous rods of 2 mm diameter can be obtained over a wide
composition range. With the addition of Cu, the free volume changes from 0.09 to 0.18%, and the fracture plastic strain changes from 1 to
12.4%. However, there is no strong correlation between the free volume and the compressive plasticity. In addition to the free volume content,
other factors such as the size, shape and distribution of the free volume sites may also affect the plasticity of BMGs.
[doi:10.2320/matertrans.M2013241]
(Received June 26, 2013; Accepted September 19, 2013; Published November 15, 2013)
Keywords: bulk metallic glass, structural relaxation, free volume, plasticity
1.
Introduction
Due to the limited room-temperature ductility of bulk
metallic glasses (BMGs), tremendous efforts have been paid
over the last decades to develop BMGs with enhanced
plasticity. A number of BMG systems with improved
ductility have been developed, including ZrTaCuNiAl,1)
ZrCuNiAl,2) ZrCuAgAl,3) ZrPdFeAlAg,4) CuAgZrTi,5) CuTiZrNbNiSi,6) CuZrAlY,7) TiZrCuNiBeNb,8) TiZrCuPdSn,9)
MgCuAgGd,10) NiZrNbAlTa.11) It is interesting to note that
in many of these BMGs with enhanced plasticity, an atomic
pair exists with positive enthalpy of mixing, i.e., Zr­Ta in
ZrTaCuNiAl, Cu­Ni in ZrCuNiAl, Cu­Ag in ZrCuAgAl and
CuAgZrTi, Fe­Ag in ZrPdFeAlAg, Zr­Nb in CuTiZrNbNiSi
and TiZrCuNiBeNb, Zr­Y in CuZrAlY, Cu­Sn in TiZrCuPdSn. Park et al. considered that the addition of alloying
elements with positive heat of mixing can induce atomicscale local inhomogeneities or fluctuations in the local free
volume distribution.6) When the constituent elements are not
mixed uniformly, the atomic-packing density becomes lower,
and thereby an open structure with more free volume is
obtained.12) It was pointed out that a large amount of free
volume will enhance the plasticity of BMGs.13,14) However,
the plasticity of BMGs usually does not increase monotonously with the addition of an element having positive heat
of mixing with the constituent elements of the alloy. More
work is therefore needed to examine the relationship and the
role of free volume on the plasticity of BMGs of different
compositions.
In the present work, the effect of alloying Cu, which has
a positive heat of mixing with Co, on the free volume and
plasticity of the Zr56Co28Al16 BMG was investigated. The
reasons for selecting this alloy and the alloying element for
the present study are: (1) Zr56Co28Al16 has a large glass
forming ability (GFA);15) (2) Cu and Co have a moderate
degree of positive heat of mixing of +6 kJ/mol,16) which is
a typical value of the heat of mixing between the alloying
element and the constituent element of the base alloy; (3) Cu
+
Corresponding author, E-mail: [email protected]
and Co have a similar atomic radius (atomic radius of Cu
and Co are 0.128 and 0.125 nm respectively17)). With the
moderate degree of positive heat of mixing between Cu and
Co and the similarity in their atomic radii, it is expected that
the substitution of Co with Cu can increase the amount of
free volume for the Zr56Co28Al16 BMG, without drastically
reducing its glass forming ability.
2.
Experimental
2.1 Sample preparation and characterization
Alloy ingots with nominal compositions of Zr56Co28¹xCuxAl16 (x = 0, 3, 6, 12, 15, 21, 28) were prepared from high
purity elements (purity > 99.8 mass%) by arc melting the
mixture of the elements in a Ti-gettered high purity argon
atmosphere. Cylindrical alloy rods of 2 mm diameter were
prepared from these ingots by water-cooled copper mold
casting. The structures of the alloys were examined by X-ray
diffraction (XRD, Bruker D8 Advanced) with monochromated Cu K¡ radiation and a transmission electron microscope (TEM, JEOL JEM 2010F). The TEM specimens were
prepared by mechanical grinding, followed by ion milling
with a Gatan 691 precision ion polishing system operating at
5 keV. Uniaxial compression tests were carried out on an
MTS 810 material testing system at a strain rate of 10¹4 s¹1
using cylindrical rods with a diameter of 2 mm and a length
of 4 mm. Both ends of the rods were carefully polished to
ensure parallelism. The deformation of the specimen was
measured by a calibrated extensometer. At least three samples
for each alloy composition were tested to check the
reproducibility of the results. The lateral surfaces of the
compression samples after fracture were examined by
scanning electron microscopy (JEOL JSM 6490).
2.2 Determination of free volumes
Free volumes of the Zr56Co28¹xCuxAl16 (x = 0, 3, 6, 12,
15, 21, 28) BMGs were quantitatively determined from the
structure relaxation enthalpy before glass transition from
isochronal differential scanning calorimetry (DSC) tests
according to eq. (1):18,19)
2210
Z. Liu, K. C. Chan and L. Liu
Z
H ¼ ¢vf ¼
Cp dT
ð1Þ
where ¢ is a constant. The physical meaning of ¢ is
analogous to the molar formation energy of free volume in
metallic glasses. For Zr-based Zr55Cu30Al10Ni5, Zr45.0Cu39.3Al7.0Ag8.7 and Zr44Ti11Ni10Cu10Be25 BMGs, the value of
¢ was reported to be 552,18) 64220) and 62321) kJ/mol
respectively, which are very close to each other. "Cp is the
specific heat difference between the metallic glass for which
the free volume is to be determined and the reference state.
Structurally relaxed metallic glasses after a long annealing
time are often used as the reference state.18,21) However, since
the degree of structural relaxation depends logarithmically
on time, the ideal reference state can never be achieved in
practice.22) Furthermore, during a long annealing time, partial
crystallization may also occur. These factors will increase
the uncertainty of the free volume value determined by the
calorimetric method. The fully-crystallized alloy of metallic
glass was taken as the reference state in the present work
since this is a state which is more definite and easily
attainable. Although the final crystallization product may
have varied grain sizes, depending on the crystallization
temperature and time, little uncertainty will be introduced
from the fully-crystallized reference state since Sun et al.
pointed out that there is no significant difference between the
specific heat of the as-crystallized nanocrystalline alloy and
the coarse-grained polycrystalline state.23)
The measurements were carried out on a differential
scanning calorimeter (DSC, Perkin Elmer DSC7) under a
constant nitrogen flow. Each measurement involves four
successive runs: an empty aluminum pan, a standard
synthetic sapphire disk of about 10 mg in weight, the ascast BMG and the fully crystallized BMG. The specific heat
capacity of the sample was then calculated from the
following formula:24,25)
Cp ðT Þs ¼ Cp ðT Þss Ds Wss
Dss Ws
ð2Þ
where Cp(T)s is specific heat capacity of the sample, Cp(T)ss is
specific heat capacity of the sapphire standard, Ds is the heat
flow difference between the empty specimen holder and the
specimen in the DSC thermal curves at a given temperature,
Dss is heat flow difference between the empty specimen
holder and the sapphire standard in the DSC thermal curves
at a given temperature, Ws is mass of the specimen and Wss is
the mass of the sapphire standard. For each run, the sample
was first kept at 30°C for 2 min, then heated to 580°C at a
heating rate of 80°C/min and finally held isothermally heated
at 580°C for 6 min. Samples of about 15 mg were used. Three
measurements in total were done for each alloy. All the
measurements for different alloys were arranged in a random
order to eliminate the possible influence from the instability
of the equipment.
3.
Results and Discussion
Figure 1 shows the XRD patterns of the as-cast
Zr56Co28¹xCuxAl16 (x = 0, 3, 6, 12, 15, 21, 28) alloys of
diameter 2 mm. All the patterns show only broad diffraction
Fig. 1 XRD patterns of the as-cast Zr56Co28¹xCuxAl16 (x = 0, 3, 6, 12, 15,
21, 28) alloy rods with a diameter of 2 mm.
maxima without any sharp diffraction peaks, indicating
the amorphous nature of the alloys. High-resolution TEM
(HRTEM) and the corresponding selected area electron
diffraction (SAED) patterns of the alloys x = 0, 3 and 15
are shown in Fig. 2. The SAED patterns exhibit only diffuse
halo rings, further confirming the glassy structure of the
alloys. All the HRTEM images show uniform featureless
contrast and no inhomogeneity can be observed. From the
results of XRD and TEM, it is found that all alloys in the
Zr56Co28¹xCuxAl16 (x = 0, 3, 6, 12, 15, 21, 28) system can be
cast into a fully amorphous rod of 2 mm in diameter. The
base alloy Zr56Co28Al16 was reported to have a critical
casting diameter of 18 mm.15) Although the exact critical
casting diameters for the other alloys in this system were not
determined, the addition of Cu in substitution for Co over
a wide composition range in the Zr56Co28¹xCuxAl16 system
seems not to cause much deterioration of the GFA since GFA
is strongly dependent on the composition.26) It is interesting
to note that for the CuZrYAl alloy system in which the heat
of mixing of the Zr­Y binary pair is +9 kJ/mol16) and the
atomic radii for Zr and Y are 0.160 and 0.180 nm
respectively,17) the glass forming ability drastically drops
from 10 mm of Cu46Zr42Y5Al727) to less than 50 µm of
Cu46Zr22Y25Al77) with the substitution of Zr with Y. It is thus
favorable to choose an alloying element having similar
atomic radius and moderate degree of positive heat of mixing
with the element to be substituted in the base alloy in order
not to significantly degrade the glass forming ability of the
base alloy.
Figure 3 shows the uniaxial compressive stress­strain
curves of the Zr56Co28¹xCuxAl16 BMG rods with x = 0, 3,
6, 12, 15, 21, 28. All the alloys exhibit a yield strength
of 1780­1980 MPa and an elastic limit of 1.9­2.1%. For
the base alloy Zr56Co28Al16, it has a fracture strength of
2078 MPa and a fracture plastic strain of 5.8%. With the
addition of 3 and 6 at% Cu in substitution for Co, both the
fracture strength and the fracture plastic strain increase.
The alloy with 3 at% Cu exhibits the highest strength of
2241 MPa and the largest fracture plastic strain of 12.4%.
Further addition of Cu leads to a decrease in both the strength
and the fracture plastic strain. When Co is completely
substituted by Cu, Zr56Co28Al16 exhibits the lowest fracture
Effect of Alloying Elements with Positive Heat of Mixing on the Free Volume and Compressive Plasticity in ZrCoCuAl Bulk Metallic Glasses 2211
(a)
(b)
5 nm
5 nm
(c)
5 nm
Fig. 2 High-resolution TEM micrographs and the corresponding selected area electron diffraction images of Zr56Co28¹xCuxAl16 alloys.
(a) x = 0, (b) x = 3, (c) x = 15.
Fig. 3 Representative uniaxial compressive stress­strain curves of the
Zr56Co28¹xCuxAl16 (x = 0, 3, 6, 12, 15, 21, 28) BMG rods with diameter
of 2 mm.
plastic deformation of 1%. The SEM lateral surface
morphologies of the fractured Zr56Co28¹xCuxAl16 (x = 0, 3,
15, 28) BMGs are shown in Fig. 4. Shear bands can be
observed in all the fractured samples. It also reveals that the
BMG with x = 3 has more intersecting and branched shear
bands, which is consistent with the result of the compression
test.
The apparent specific heat capacity (Cp) curves of the ascast and fully crystallized Zr56Co28¹xCuxAl16 (x = 0, 3, 6, 12,
15, 21, 28) alloys are depicted in Fig. 5. For the as-cast BMG
and its fully crystallized alloy, both their Cp values are nearly
identical and increase slightly when the temperature is below
about 500 K. With increasing temperature, the Cp of the ascast BMG drops and its value becomes smaller than that of
the fully crystallized alloy. The point at which the Cp of
the as-cast BMG falls below the fully crystallized alloy is
defined as the onset temperature of structural relaxation
(Tronset). With further increase of the temperature above the
onset relaxation temperature, the Cp of the fully crystallized
alloy continues to increase slightly, whereas the Cp of the ascast BMG decreases more rapidly before increasing. When
the value of the as-cast BMG becomes identical with the
fully crystallized alloy, the temperature is defined as the end
temperature of the structural relaxation (Trend). The Cp of the
as-cast BMG will then increase abruptly above Trend, which
corresponds to glass transition into the supercooled liquid
region. The Tronset, Trend and glass transition onset temperature
(Tgonset) for the as-cast Zr56Co28¹xCuxAl16 (x = 0, 3, 6, 12, 15,
21, 28) BMGs are shown in Fig. 6. Tronset is found to change
between ³450 K and ³550, whereas Trend and Tgonset are
about the same for all the x values and they decrease slightly
with the addition of Cu. Hu et al. pointed out that the Trend
2212
Z. Liu, K. C. Chan and L. Liu
(a)
(b)
200 µm
(c)
200 µm
(d)
200 µm
200 µm
Fig. 4 SEM images of the fractured Zr56Co28¹xCuxAl16 alloys from the lateral view. (a) x = 0, (b) x = 3, (c) x = 15 and (d) x = 28.
corresponds to the temperature at the end of the structural
relaxation as well as the onset of glass transition Tgonset,28)
which is consistent with the results of the current study. The
structural relaxation enthalpies ("H) for the as-cast BMGs
are calculated by integrating the specific heat difference
between the as-cast BMG and the fully crystallized alloy over
the temperature range of Tronset and Trend. The relaxation
enthalpies fall in the range of 475.9­974.4 J/mol. From these
"H values, the free volume ("vf ) was derived from eq. (1),
with the constant ¢ chosen as 552 kJ/mol.18) Figure 6(b)
shows the free volume and the fracture plastic strain of the ascast Zr56Co28¹xCuxAl16 (x = 0, 3, 6, 12, 15, 21, 28) BMGs.
The free volumes of the as-cast Zr56Co28¹xCuxAl16 (x = 0, 3,
6, 12, 15, 21, 28) BMGs are in the range of 0.09­0.18%,
which is in good agreement with the reported values of 0.04­
0.25% of other BMG systems.18,21,28) Figure 6(b) also reveals
that the free volume does not have a strong correlation with
the amount of Cu, which has a positive heat of mixing with
Co. When the fracture plastic strain-composition curve as
shown in Fig. 6(b) is also considered, it demonstrates that for
BMGs with different compositions it is not necessary that
there is strong correlation between the free volume content
and the fracture plastic strain, which differs from the results
reported by Xie et al., in which there was a qualitative
correlation between the free volume and compressive plastic
strain in BMGs.14) In the Zr56Co28¹xCuxAl16 alloy system,
due to solute­solute avoidance29) and a repulsive interaction
between the Co and Cu atoms from the positive heat of
mixing between Co and Cu, the Co and Cu atoms will
principally coordinate with Zr and/or Al atoms. Encaged in a
coordinate shell composed of Zr and Al atoms, Co atoms will
not be affected due to the existence of Cu atoms. As a result,
the addition of Cu, having a positive heat of mixing with Co,
will not reduce the packing density and increase the free
volume of the Zr56Co28¹xCuxAl16 alloys, as may be supposed.
With regard to the conclusion that no positive correlation can
be found between the free volume content and the plasticity,
it can be understood from the fact that the plasticity of
metallic glasses not only depends on the total free volume
content, but also is influenced by factors such as the size,
shape and spatial distribution of the free volume sites.30,31)
Besides the free volume, Yokoyama et al. pointed out that the
plasticity and toughness of metallic glass is also associated
with the degree of metallic bonding.32) For metallic glasses
with different compositions, even through their free volumes
are identical, they may have a distinct degree of metallic
bonding and dissimilar plasticity. In metallic glass with the
same chemical composition but different processing conditions such as quenching rate,33,34) annealing35,36) and
mechanical pre-deformation,37,38) a good correlation between
the free volume content and the plasticity can usually be
found. Contrasting with the findings in this work, it can
Effect of Alloying Elements with Positive Heat of Mixing on the Free Volume and Compressive Plasticity in ZrCoCuAl Bulk Metallic Glasses 2213
Fig. 5
Apparent specific heat capacity curves of the as-cast and fully crystallized Zr56Co28¹xCuxAl16 (x = 0, 3, 6, 12, 15, 21, 28) alloys.
be concluded that although free volume in metallic glass
changes with chemical composition, free volume alone
cannot fully describe the influence of the composition on
the structure and the mechanical properties of metallic glasses
with different compositions. In some circumstances where
the chemical effect does not exist, the free volume can be
used as a useful parameter to bridge the structure and
properties of metallic glasses.
4.
Conclusion
Effect of addition of an element having positive heat of
mixing with the constituent elements of metallic glass on the
free volume and compressive plasticity was investigated in
the Zr56Co28¹xCuxAl16 (x = 0, 3, 6, 12, 15, 21, 28) alloy
system. Fully amorphous alloy rods with a diameter of
2 mm can be prepared over the entire composition range
2214
Z. Liu, K. C. Chan and L. Liu
(a)
(b)
Fig. 6 (a) Structural relaxation onset, end temperature and the glass
transition temperature of the as-cast Zr56Co28¹xCuxAl16 (x = 0, 3, 6, 12,
15, 21, 28) alloys. (b) Free volume and plastic strain of the as-cast
Zr56Co28¹xCuxAl16 (x = 0, 3, 6, 12, 15, 21, 28) alloys.
by copper mold casting. The compressive plasticity initially
increases and then decreases with the addition of Cu, with
the alloy of x = 3 exhibiting the highest strength of
2241 MPa and the largest fracture plastic strain of 12.4%.
The specific heat capacities for the as-cast metallic glass
and its fully-crystallized alloy are nearly the same at
relatively lower temperatures. Above ³450­530 K during
isochronal heating at 80 K/min, the apparent specific heat of
the as-cast metallic glass falls below the fully-crystallized
alloy and structural relaxation process occurs, which completes at the glass transition temperature. The free volume
calculated from the structural relaxation enthalpy ranges
from 0.09 to 0.18%. No definite correlations among the
addition of an element with positive heat of mixing, the
free volume and the compressive plasticity can be found in
the Zr56Co28¹xCuxAl16 (x = 0, 3, 6, 12, 15, 21, 28) alloy
system.
Acknowledgments
This research was financially supported by the grants from
the Research Grants Council of The Hong Kong Special
Administration Region (Project Nos.: PolyU511510 and
PolyU511108).
REFERENCES
1) L.-Q. Xing, Y. Li, K. T. Ramesh, J. Li and T. C. Hufnagel: Phys. Rev. B
64 (2001) 180201.
2) Y. H. Liu, G. Wang, R. J. Wang, D. Q. Zhao, M. X. Pan and W. H.
Wang: Science 315 (2007) 1385­1388.
3) Q. K. Jiang, X. D. Wang, X. P. Nie, G. Q. Zhang, H. Ma, H.-J. Fecht, J.
Bendnarcik, H. Franz, Y. G. Liu, Q. P. Cao and J. Z. Jiang: Acta Mater.
56 (2008) 1785­1796.
4) Z. Liu, K. C. Chan and L. Liu: Mater. Trans. 52 (2011) 61­67.
5) E. S. Park, H. J. Chang, D. H. Kim, T. Ohkubo and K. Hono: Scr.
Mater. 54 (2006) 1569­1573.
6) E. S. Park, D. H. Kim, T. Ohkubo and K. Hono: J. Non-Cryst. Solids
351 (2005) 1232­1238.
7) E. S. Park and D. H. Kim: Acta Mater. 54 (2006) 2597­2604.
8) S. F. Shan, Z. J. Zhan, C. Z. Fan, Y. Z. Jia, B. Q. Zhang, R. P. Liu and
W. K. Wang: Chin. Phys. Lett. 25 (2008) 4165­4167.
9) S. L. Zhu, X. M. Wang and A. Inoue: Intermetallics 16 (2008) 1031­
1035.
10) E. S. Park, J. Y. Lee and D. H. Kim: J. Mater. Res. 20 (2005) 2379­
2385.
11) J. H. Na, J. M. Park, K. H. Han, B. J. Park, W. T. Kim and D. H. Kim:
Mater. Sci. Eng. A 431 (2006) 306­310.
12) S. W. Lee, S. C. Lee, Y. C. Kim, E. Fleury and J. C. Lee: J. Mater. Res.
22 (2007) 486­492.
13) L. Y. Chen, Z. D. Fu, G. Q. Zhang, X. P. Hao, Q. K. Jiang, X. D. Wang,
Q. P. Cao, H. Franz, Y. G. Liu, H. S. Xie, S. L. Zhang, B. Y. Wang,
Y. W. Zeng and J. Z. Jiang: Phys. Rev. Lett. 100 (2008) 075501.
14) S. H. Xie, X. R. Zeng and H. X. Qian: J. Alloy. Compd. 480 (2009)
L37­L40.
15) T. Wada, F. X. Qin, X. M. Wang, M. Yoshimura and A. Inoue: J. Mater.
Res. 24 (2009) 2941­2948.
16) A. Takeuchi and A. Inoue: Mater. Trans. 46 (2005) 2817­2829.
17) O. N. Senkov and D. B. Miracle: Mater. Res. Bull. 36 (2001) 2183­
2198.
18) A. Slipenyuk and J. Eckert: Scr. Mater. 50 (2004) 39­44.
19) A. van den Beukel and J. Sietsma: Acta Metall. Mater. 38 (1990) 383­
389.
20) Y. Zhang and H. Hahn: J. Alloy. Compd. 488 (2009) 65­71.
21) Z. Evenson and R. Busch: Acta Mater. 59 (2011) 4404­4415.
22) D. B. Miracle, T. Egami, K. M. Flores and K. F. Kelton: MRS Bull. 32
(2007) 629­634.
23) N. X. Sun and K. Lu: Phys. Rev. B 54 (1996) 6058­6061.
24) American Society for Testing and Materials: ASTM E1269-05
Standard Test Method for Determining Specific Heat Capacity by
Differential Scanning Calorimetry, (2005).
25) M. J. O’Neill: Anal. Chem. 38 (1966) 1331­1336.
26) D. Ma, H. Tan, D. Wang, Y. Li and E. Ma: Appl. Phys. Lett. 86 (2005)
191906.
27) D. H. Xu, G. Duan and W. L. Johnson: Phys. Rev. Lett. 92 (2004)
245504.
28) Q. Hu, X. R. Zeng and M. W. Fu: J. Appl. Phys. 109 (2011) 053520.
29) D. B. Miracle: Acta Mater. 54 (2006) 4317­4336.
30) J. Sietsma and B. J. Thijsse: Phys. Rev. B 52 (1995) 3248­3255.
31) K. M. Flores, E. Sherer, A. Bharathula, H. Chen and Y. C. Jean: Acta
Mater. 55 (2007) 3403­3411.
32) Y. Yokoyama, T. Yamasaki, P. K. Liaw and A. Inoue: Acta Mater. 56
(2008) 6097­6108.
33) Y. J. Huang, J. Shen and J. F. Sun: Appl. Phys. Lett. 90 (2007) 081919.
34) L. Y. Chen, A. D. Setyawan, H. Kato, A. Inoue, G. Q. Zhang, J. Saida,
X. D. Wang, Q. P. Cao and J. Z. Jiang: Scr. Mater. 59 (2008) 75­78.
35) K. Mondal, T. Ohkubo, T. Toyama, Y. Nagai, M. Hasegawa and K.
Hono: Acta Mater. 56 (2008) 5329­5339.
36) F. Jiang, Y. L. Zhao, L. C. Zhang, S. B. Pan, Y. G. Zhou, L. He and
J. Sun: Adv. Eng. Mater. 11 (2009) 177­181.
37) M. S. Pham, K. W. Park, B. G. Yoo, J. I. Jang and J. C. Lee: Met.
Mater. Int. 15 (2009) 209­214.
38) Y. J. Huang, Y. Sun and J. Sun: Intermetallics 18 (2010) 2044­2050.